← Models

Model profile

Llama 2 Chinese 7B Chat

2023-07-23release date
2eval lineages

Evidence summary

Published evidence spans 2 evals and 3 of 7 behavior components. Its strongest relative result is CMoralEval (social_morality, #16 of 26); its weakest is SafetyBench (PH, #21 of 21).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

Published eval results

Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.

Eval / sub-evalRankValueDistributionBetterSource
CMoralEvalfamilial_morality#18 / 260.37↑ higherSource ↗official
CMoralEvalinternet_ethics#18 / 260.35↑ higherSource ↗official
CMoralEvalpersonal_morality#17 / 260.36↑ higherSource ↗official
CMoralEvalprofessional_ethics#17 / 260.37↑ higherSource ↗official
CMoralEvalsocial_morality#16 / 260.37↑ higherSource ↗official
SafetyBenchEM#21 / 2143.4↑ higherSource ↗official
SafetyBenchIA#21 / 2153.5↑ higherSource ↗official
SafetyBenchMH#20 / 2161.7↑ higherSource ↗official
SafetyBenchOFF#18 / 2148.9↑ higherSource ↗official
SafetyBenchPH#21 / 2143↑ higherSource ↗official
SafetyBenchPP#20 / 2157.6↑ higherSource ↗official
SafetyBenchUB#13 / 2161.3↑ higherSource ↗official